Deep Life Sci

A LangChain Deep Agent assistant for life scientists and chemists. The agent searches PubMed, PMC full texts, and ClinicalTrials.gov to answer questions and generate figures.

Quickstart

1. Get the code

In the terminal:

git clone https://github.com/mcunningham1440/deep-life-sci.git

cd deep-life-sci

Needs git, which setup also uses to fetch the chat UI.

2. Get uv

uv is a widely-used package manager for Python that allows the setup script to install the necessary libraries.

On macOS / Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

On Windows:

irm https://astral.sh/uv/install.ps1 | iex

3. Run the setup script

uv run scripts/setup.py

4. Add your API key

You need a LangSmith account. Setup will prompt you to add your LANGSMITH_API_KEY.

Model calls go through the LangSmith LLM gateway, so your workspace also needs the provider key behind them, added once under Settings → Integrations → Provider Secrets as OPENAI_API_KEY and/or ANTHROPIC_API_KEY. Add whichever providers the models you run use.

ROOT_MODEL=claude-sonnet-5 in front of either command below swaps a model (see research_agent/models.py).

5. Run

uv run scripts/dev.py                        # opens the chat UI in your browser (recommended)

# or

uv run agent "which papers base-edit PCSK9?" # runs headlessly in CLI

Ctrl-C to stop the running server.

Layout

scripts/              setup.py (one-time setup), dev.py (chat stack), build_snapshot.py
research_agent/       the agent: assembly, entry points, tools, prompts, middleware
evals/                LangSmith datasets + evaluators
ui/                   artifact components rendered by the chat frontend
docs/                 design notes and demo questions

Notes

The UI is a clone of agent-chat-ui in .chat-ui/. It needs Node 20+, the one thing setup won't install for you: everything else finishes without it, so install Node and re-run to add the UI.

Full text journal articles are only available if present in PMC's open-access subset. Only abstracts are available for paywalled papers.

MIT Licensed